An innovative AI platform utilizing Mel-spectrograms and convolutional neural networks (CNN) was developed to objectively assess the severity of unilateral vocal cord paralysis (UVCP). The system, named TripleConvNet, was trained on voice samples from over 400 subjects, accurately distinguishing levels of vocal fold compensation. This represents a significant advancement in otolaryngology diagnostics by providing a non-invasive, standardized tool for UVCP evaluation.